The power of affective pedagogical agent and self-explanation in computer-based learning

The goal of the current study was to investigate the effects of affective pedagogical agent (PA) and student-generated explanation on learners' affective processing, visual attention, and learning outcomes during a video lesson. In a 2 (affective pedagogical agent: positive PA vs. neutral PA) × 2 (generative learning strategy: self-explanation vs. re-study) between-subjects design, participants watched an animation on synaptic transmission from a multimedia lesson that was either delivered by a positive PA or a neutral PA. After watching the video, students either generated a verbal explanation or rewatched the video. All students’ eye movements were tracked during learning and then took learning outcome tests. The results showed that: (a) students who learned with a positive PA paid more attention to relevant elements of the materials and performed better on learning outcomes than those who learned with a neutral PA; (b) self-explanation facilitated visual processing and learning outcomes; and (c) positive PA accompanying self-explanation increased intrinsic motivation and resulted in best learning performance. The findings support the use of a positive PA to design multimedia learning environments and the use of self-explanation strategy to engage students in generative processing activities. Moreover, students benefit most from positive PA and then generating an oral explanation, which demonstrates the importance of combining instructional design features and learners-generated learning activity in computer-based learning environments. •We tested the effects of pairing affective pedagogical agent (PA) with self-explanation in a video lesson.•Positive PA facilitated visual processing and learning outcomes.•Self-explanation activity helped learners focus on relevant elements and improve learning performance.•Students generated high-quality explanations in the positive PA condition.•Students benefited most from aligning positive PA with self-generated oral explanation.